Small Sample Learning in Big Data Era
arXiv:1808.04572
Abstract
As a promising area in artificial intelligence, a new learning paradigm, called Small Sample Learning (SSL), has been attracting prominent research attention in the recent years. In this paper, we aim to present a survey to comprehensively introduce the current techniques proposed on this topic. Specifically, current SSL techniques can be mainly divided into two categories. The first category of SSL approaches can be called "concept learning", which emphasizes learning new concepts from only few related observations. The purpose is mainly to simulate human learning behaviors like recognition, generation, imagination, synthesis and analysis. The second category is called "experience learning", which usually co-exists with the large sample learning manner of conventional machine learning. This category mainly focuses on learning with insufficient samples, and can also be called small data learning in some literatures. More extensive surveys on both categories of SSL techniques are introduced and some neuroscience evidences are provided to clarify the rationality of the entire SSL regime, and the relationship with human learning process. Some discussions on the main challenges and possible future research directions along this line are also presented.
arXiv comment76 pages, 15 figures, survey of small sample learning
References in corpus (69)
- Auto-Encoding Variational Bayes
- Distilling the Knowledge in a Neural Network
- Sequence to Sequence Learning with Neural Networks
- Convolutional LSTM Network: A Machine Learning Approach for Precipitation Nowcasting
- Model-Agnostic Meta-Learning for Fast Adaptation of Deep Networks
- WaveNet: A Generative Model for Raw Audio
- Learning Transferable Features with Deep Adaptation Networks
- Generative Adversarial Text to Image Synthesis
- Conditional Image Synthesis With Auxiliary Classifier GANs
- Theoretical Models of Learning to Learn
- Unsupervised Domain Adaptation with Residual Transfer Networks
- A Simple Neural Attentive Meta-Learner
- Meta-SGD: Learning to Learn Quickly for Few-Shot Learning
- Data Augmentation Generative Adversarial Networks
- Snorkel: Rapid Training Data Creation with Weak Supervision
- Zero-Shot Learning by Convex Combination of Semantic Embeddings
- Deep Convolutional Inverse Graphics Network
- AutoAugment: Learning Augmentation Policies from Data
- Dynamic Memory Networks for Visual and Textual Question Answering
- RL: Fast Reinforcement Learning via Slow Reinforcement Learning
- Few-Shot Learning with Graph Neural Networks
- Decomposing Motion and Content for Natural Video Sequence Prediction
- Deep Predictive Coding Networks for Video Prediction and Unsupervised Learning
- Learning to reinforcement learn
- Like What You Like: Knowledge Distill via Neuron Selectivity Transfer
- Multi-source Transfer Learning with Convolutional Neural Networks for Lung Pattern Analysis
- Domain Adaptation for Visual Applications: A Comprehensive Survey
- Attend, Infer, Repeat: Fast Scene Understanding with Generative Models
- Learning to Remember Rare Events
- Triple Generative Adversarial Nets
- Learned D-AMP: Principled Neural Network based Compressive Image Recovery
- Few-Example Object Detection with Model Communication
- Intrinsically Motivated Goal Exploration Processes with Automatic Curriculum Learning
- Model-Free Episodic Control
- Learning to Optimize
- Continuous Adaptation via Meta-Learning in Nonstationary and Competitive Environments
- Towards a Neural Statistician
- Improvements to context based self-supervised learning
- Information-Theoretical Learning of Discriminative Clusters for Unsupervised Domain Adaptation
- Unrolled Optimization with Deep Priors
- Meta Learning Shared Hierarchies
- Zero-Shot Detection
- Meta-Learning and Universality: Deep Representations and Gradient Descent can Approximate any Learning Algorithm
- SwellShark: A Generative Model for Biomedical Named Entity Recognition without Labeled Data
- LSTD: A Low-Shot Transfer Detector for Object Detection
- What Objective Does Self-paced Learning Indeed Optimize?
- Sample-to-Sample Correspondence for Unsupervised Domain Adaptation
- Learning Deep Representations of Medical Images using Siamese CNNs with Application to Content-Based Image Retrieval
- Gaussian Prototypical Networks for Few-Shot Learning on Omniglot
- Zero-Shot Learning by Generating Pseudo Feature Representations
- Adapted Deep Embeddings: A Synthesis of Methods for -Shot Inductive Transfer Learning
- Generative Models of Visually Grounded Imagination
- Metacontrol for Adaptive Imagination-Based Optimization
- Generative Temporal Models with Memory
- Zero-Shot Learning posed as a Missing Data Problem
- Semi-supervised Zero-Shot Learning by a Clustering-based Approach
- Deep Triplet Ranking Networks for One-Shot Recognition
- Computational Theories of Curiosity-Driven Learning
- Deep Matching Autoencoders
- Lung Cancer Screening Using Adaptive Memory-Augmented Recurrent Networks
- Artwork Identification from Wearable Camera Images for Enhancing Experience of Museum Audiences
- Machine Learning with World Knowledge: The Position and Survey
- MAVOT: Memory-Augmented Video Object Tracking
- Object-Level Representation Learning for Few-Shot Image Classification
- EZLearn: Exploiting Organic Supervision in Large-Scale Data Annotation
- Zero-Shot Learning via Category-Specific Visual-Semantic Mapping
- Dynamic Input Structure and Network Assembly for Few-Shot Learning
- Labeled Memory Networks for Online Model Adaptation
- Meta-Learning via Feature-Label Memory Network
Cited by in corpus (15)
- Generalizing from a Few Examples: A Survey on Few-Shot Learning
- Meta-Weight-Net: Learning an Explicit Mapping For Sample Weighting
- A Two-Stage Approach to Few-Shot Learning for Image Recognition
- Reinforcement Learning in Healthcare: A Survey
- OSLNet: Deep Small-Sample Classification with an Orthogonal Softmax Layer
- A Survey on Deep Learning of Small Sample in Biomedical Image Analysis
- Meta Transition Adaptation for Robust Deep Learning with Noisy Labels
- Learning Adaptive Loss for Robust Learning with Noisy Labels
- A Concise Review of Recent Few-shot Meta-learning Methods
- A Very Brief Introduction to Machine Learning With Applications to Communication Systems
- Sample-based Regularization: A Transfer Learning Strategy Toward Better Generalization
- Select-ProtoNet: Learning to Select for Few-Shot Disease Subtype Prediction
- What Can Knowledge Bring to Machine Learning? -- A Survey of Low-shot Learning for Structured Data
- Coarse-To-Fine Incremental Few-Shot Learning
- Multi-label Classification for Automatic Tag Prediction in the Context of Programming Challenges